+1 to this - @goccy is this repo actively maintained still?

+1 actively need this as well as the default streaming method uses the "_default" stream and you can't use connections pool without that : Trying to enable connection pool in non-default stream.

I tried a workaround for this consisting of creating the "_default" write stream after the table but It seems like the API doesn't allow us to set the name of the stream -> OUTPUT ONLY.

var streamName = WriteStreamName.of(bQProjectId, datasetId, tableId, "_default").toString();
CreateWriteStreamRequest request =
    CreateWriteStreamRequest.newBuilder()
        .setParent(TableName.of(bQProjectId, datasetId, tableId).toString())
        .setWriteStream(
            WriteStream.newBuilder()
                .setType(WriteStream.Type.COMMITTED)
                .setName(streamName)
                .build())
        .build();
WriteStream response = bigQueryWriteClient.createWriteStream(request);
System.out.println("WriteStream created " + response);

With the following code, the stream created have a randomly generated name.

+1 i need this with go client.

Here's my example Go code using the managedwriter package to request the bigquery emulator:

	ctx := context.Background()
	grpcCtx, _ := context.WithTimeout(ctx, 1*time.Second)
	conn, _ := grpc.DialContext(grpcCtx, "localhost:9060",
		grpc.WithTransportCredentials(insecure.NewCredentials()),
		grpc.WithIdleTimeout(1*time.Second),
	)
	client, err := managedwriter.NewClient(ctx, projectID,
		option.WithGRPCConn(conn),
		option.WithoutAuthentication(),
	) // OK

	pb := &v1.pb{}
	descriptorProto, err := adapt.NormalizeDescriptor(pb.ProtoReflect().Descriptor())
	tableName := fmt.Sprintf("projects/%s/datasets/%s/tables/%s", projectID, datasetID, "{table_name}")
	stream, err := client.NewManagedStream(ctx,
		managedwriter.WithDestinationTable(tableName),
		managedwriter.WithSchemaDescriptor(descriptorProto),
		managedwriter.WithAppendRowsCallOption(gax.WithTimeout(1*time.Second)),
	) // return error
	
	// rpc error: code = Unknown desc = failed to find stream from projects/{project}/datasets/{dataset}/tables/{table}/streams/_default

When calling NewManagedStream() without specifying the stream type option, it internally invokes getWriteStream with the stream name argument projects/{project}/datasets/{dataset}/tables/{table}/streams/_default.

IMO, the API should automatically create a _default stream for each table if it doesn't exist at the time of the call, or alternatively, prepare a _default stream for all tables during the initialization phase.

Here's a quick fix I tried as an alternative, but I'm not sure if it's the right approach: araj-dev#1

+1 for default straem support

@goccy added test case for default stream

@goccy Could you please merge this PR into the main branch? I also need it.

@MatanLevy I would like to check the CI results, so could you push an empty commit to trigger the CI ?

@goccy pushed now an empty commit

Thank you for your contribution !!! LGTM 👍